Darwin Bautista

@PSanni @siddagra @bmusq As of commit b290950dad5a3dceb574cbc2d902765e1496ace2, finetuning is now officially supported. `checkpoint` parameter of `test.py` and `read.py` has been changed accordingly. Now you can do: ```shell # Finetuning ./train.py...

@PSanni > Yes, i did. But the problem is with size of embedding layer, which is 95 (char in 94_full.yaml). Therefore, including any additional characters cause mismatch error. So, i...

> @baudm I have try to recognize Image(contain the sentence input). I know that your model now use for word level. My question is: Does model can train input image(sentence)?...

Tried it both on Colab (new notebook) and local, and it works as expected: For Colab, make sure you remove old cached files in `/root/.cache/torch/hub/baudm_parseq_main`

Tried it just now. Was able to export to ONNX using `torch.onnx.export(parseq, dummy_input, 'parseq.onnx', opset_version=14)`. Not really familiar yet with ONNX so I can't verify if the exported model works...

```python import torch parseq = torch.hub.load('baudm/parseq', 'parseq', pretrained=True).eval() dummy_input = torch.rand(1, 3, *parseq.hparams.img_size) # (1, 3, 32, 128) by default # To ONNX parseq.to_onnx('parseq.onnx', dummy_input, opset_version=14) # opset v14 or...

@siddagra > It is doing relatively fine for single line plates but **gets many errors on the double line ones**. STR as a task is generally limited to single-line text...

@siddagra > I did not want to do this as it adds more inference time, and most text spotters are somewhat inaccurate on my dataset, and we do not have...

@siddagra > Hey, I am unable to load https://github.com/baudm/parseq/releases/download/v1.0.0/parseq_small_patch16_224-fcf06f5a.pt (the weights for 224x224) as the config file and/or state dict are not available in the checkpoint. Any way you could...

This is a good primer on SWA: https://pytorch.org/blog/pytorch-1.6-now-includes-stochastic-weight-averaging/ SWA improves generalization. I decided to keep it in the training pipeline since it significantly improved CRNN performance (the difference is less...